AI Architect

About the Company: Nuaav is a boutique technology consulting firm focused on delivering innovative, scalable, and secure data engineering and AI solutions. We partner with corporate clients to drive impactful change through specialized expertise in product engineering, data systems, AI/ML, and user experience. Our services emphasize personalized attention, agility, and quality, enabling clients to confidently transform and grow in evolving business environments.

About the Role: We are seeking a Senior Agentic AI Engineer to design, build and scale production AI agent systems. You will own the agent harness, evaluation framework, guardrails and routing layer, with security built in from day one. You will work closely with product, platform, security and risk teams to take agents from prototype to production.

Responsibilities: Design and build the agentic harness: agent loops, tool/function calling, planning, memory, multi-agent orchestration and human-in-the-loop workflows. Develop and own evaluation frameworks (offline and online) measuring task success, tool-call accuracy, groundedness, cost and latency; integrate evals into CI/CD as release gates. Implement guardrails for input/output safety, prompt-injection and jailbreak defence, PII redaction and tool-level permissions. Build intelligent routing across LLMs, agents and tools based on intent, complexity, cost, latency and risk, with fallback and failover. Apply AI security best practices: threat modelling (OWASP Top 10 for LLMs, MITRE ATLAS), leastprivilege access, secrets management, sandboxing and audit logging. Deploy and operate agents in production with full observability, tracing and monitoring. Mentor engineers and define engineering standards for agent development.

Qualifications: Bachelor's or Master's degree in Computer Science, Engineering or a related field, or equivalent experience.5+ years of software engineering experience, including 2+ years building LLM or agentic applications in production. Proven hands-on experience with agent harnesses, evals, guardrails and model routing. Experience with agent frameworks and protocols such as LangGraph, OpenAI Agents SDK, Claude Agent SDK, CrewAI, AutoGen or MCP.Experience with eval and guardrail tooling such as LangSmith, Langfuse, Braintrust, promptfoo, NeMo Guardrails or Guardrails AI.Strong security background: threat modelling, IAM, secure API design and data protection. Strong proficiency in Python; experience with RAG, vector databases, Docker/Kubernetes and AWS, Azure or GCP.Excellent communication skills and the ability to explain AI risks to non-technical stakeholders.

Required Skills: Agent harnesses, evaluation frameworks, guardrails and model routing. Agent frameworks and protocols such as LangGraph, OpenAI Agents SDK, Claude Agent SDK, CrewAI, AutoGen or MCP.Eval and guardrail tooling such as LangSmith, Langfuse, Braintrust, promptfoo, NeMo Guardrails or Guardrails AI.Security best practices including threat modelling, IAM, secure API design and data protection. Python, RAG, vector databases, Docker/Kubernetes and AWS, Azure or GCP.Communication skills to explain AI risks to non-technical stakeholders.

Preferred Skills: Experience in banking, financial services or other regulated industries (e.g. KYC/AML, fraud detection, customer service, operations).Familiarity with AI governance and model risk frameworks such as SR 11-7, EU AI Act, NIST AI RMF and ISO/IEC 42001.Knowledge of PCI DSS, SOC 2 or ISO 27001 compliance requirements. xhqgsiq Security or cloud certifications (e.g. CISSP, OSCP, AWS/Azure/GCP).